The proposed study will investigate the effects of body mass index (BMI) on a range of individual economic outcomes. It is critically important that we understand whether adult BMI causally changes such outcomes as average hourly wages, family income, and wealth. If such causal effects exist and are nontrivial in magnitude, they provide an additional reason for combating adult obesity and eliminating barriers to the economic success of obese workers. After all, economic resources can affect the well-being of adults and enable parents to invest in the health, education, and overall well-being of their children long before the direct health risks associated with obesity become problematic. Causal links between BMI and economic outcomes have not received the research attention they deserve. The proposed study will have three specific aims. The first aim is to isolate causal effects of BMI on economic outcomes from the concomitant effects of marital transitions. Existing evidence that obesity lowers the wages of women (especially whites) but not men could simply reflect a marriage effect, given that men and women tend to gain weight upon marrying, while men's (women's) wages tend to increase (decrease) upon marrying. The proposed project will be the first to systematically isolate effects of BMI on economic outcomes from marriage effects. The second aim is to identify separate effects of weight gain and weight loss on economic outcomes. This analysis has policy implications insofar as it will suggest whether efforts should focus on preventing weight gain or promoting weight loss. The proposed project will be the first to explore whether entry into and exit from obesity have differential effects. The third aim is to identify the causal effect of BMI on four increasingly broad economic outcomes: hourly wages, annual earnings, family income, and family wealth. By investigating how BMI effects change across outcomes, the proposed project will provide new evidence on the mechanisms by which BMI influences economic well-being. To meet these specific aims, the proposed project will use large, representative samples of individuals from the 1979 National Longitudinal Survey of Youth; respondents are followed over 25 years of adulthood, from ages 22-29 to 46-53. Outcome models will be estimated using first- difference estimation. This strategy exploits life-cycle variation in body weight and other factors, and facilitates separate identification of (a) BMI effects versus the effects of transitions into and out of marital unions; and (b) weight gain and weight loss. Separate analyses will be conducted for men and women, and for blacks, whites and Hispanics of each sex.